Bean genotypes

It evaluated 70 genotypes of opaque black bean from the Mesoamerican collection, which included 20 commercial varieties, most developed by the National Institute of Forestry, Agriculture and Livestock Research (INIFAP), and 50 advanced recombinant lines (LR) derived from crosses ‛Black Papaloapan’/SEN-46, ‛Black Citlali’/XRAV-187-3 and ‛Jamapa plus’/XRAV-187-3 (Table 1).

Viral strains

The phenotyping of resistance to BCMV and BCMNV was carried out with strain BCMNV NL-3, preserved in bean seeds of the variety Michelite 62, and provided by the Plant-Virus Interactions Laboratory of the Center for Research and Advanced Studies of the Polytechnic Institute National (CINVESTAV), in Irapuato, Mexico; to phenotyping resistance to BGYMV, components A and B (accession number to GenBank AF173555 and AF173556) of strain BGYMV-MX, provided by Dr. Robert L. Gilbertson of the Department of Plant Pathology of the University of California, Davis, USA.

Phenotyping resistance to BCMV, BCMNV and BGYMV

All the resistance evaluations were carried out under confined greenhouse conditions in the INIFAP facilities. The phenotyping of resistance to BCMV and BCMNV was performed in the Bajio Experimental Field located at 20° 34’ 48.75’’ north latitude 100° 49’ 16.490’’ west longitude, at an altitude of 1767 m, and that of BGYMV in the Campo Experimental Center of Chiapas located at 16° 46’ 48’’ north latitude 93° 24’ 21’’ west longitude, at an altitude of 800 m. To phenotyping the resistance to BCMV and BCMNV, 12 plants of each genotype were inoculated mechanically.

The inoculum was obtained by macerating 1 g of the foliar tissue of the Michelite 62 variety infected with BCMNV NL-3 in 0.01 M phosphate buffer pH 7.5 (1:10 w/v); a sterile cotton gauze was moistened in the macerated solution, and the primary leaves stretched well in stage V2 were rubbed. The inoculation was performed between 7:00 and 9:00 am. The incidence of common mosaic and black root symptoms was determined visually seven days after inoculation (DDI), according to the range of typical symptoms developed in beans inoculated with BCMNV (^{Drijfhout, 1978}). Those plants that did not present symptoms were re-inoculated with the same method in the two primary leaves and the first trifoil, and the incidence of 21 DDI symptoms was determined. Genotypes that showed no symptoms after re-inoculation were considered resistant.

To phenotypeize the resistance to BGYMV, 10 plants of each genotype were agro-inoculated in stage V2 by means of syringe puncture in the first internode (^{Hou et al., 1988}). As inoculum, a 1:1 mixture of cell suspensions of the clones of BGYMV in agrobacterium pBGMXAbin (DNA-A) and pBGMXBbin (DNA-B) (^{Garrido-Ramírez et al., 2000}), previously cultivated in liquid medium LB + Kanamycin (90 mg L^-1) was used, in shaking for 48 h and analyzed by PCR to verify the permanence of both components.

The first symptoms of yellowing of ribs, golden mosaic and distortion of leaves were evaluated 15 DDI. The plants that showed some of these symptoms were conserved up to stage R8 (60 DDI) to register the development of the symptoms that included shortening of internodes, flower abortion and deformation of pods. Confirmation of infection by BGYMV was performed by PCR with the degenerate primers PAL1v1978/PAR1c496 for component A and PBLv2040/PCRc2 for component B following the procedure described by ^{Rojas et al. (1993)}.

Genotyping with molecular markers

To genotype the resistance to BCMV and BCMNV, the total DNA extraction was performed with the extraction method based on CTAB (^{Abarshi et al., 2010}), while for the BGYMV the Dellaporta method was used (^{Gilbertson et al., 1991}). The genotyping of resistance to BCMV and BCMNV was carried out with the codominant MM of CAPS ENM type, linked to the bc-3 gene, and the MMs of type SCAR SW13, linked to gene I, and SBD5, linked to the bcI ² gene, according to ^{Naderpour et al. (2010)}; ^{Melotto et al. (1996)}; ^{Miklas et al. (2000a)}, respectively. The expected amplicon size for MM SW13 and SBD5 was 650 bp and 1 400 bp, respectively.

The amplification of a 541 bp fragment with the MM ENM was confirmed by electrophoresis, and the restriction of 1 μL of the PCR product in a final volume of 10 μL was performed with 1 U of the Fau I enzyme in 1x buffer according to the manufacturer's instructions (New England Biolabs). Resistant homozygotes (eIF4E ² /eIF4E ² ) presented a fragment of 418 bp and another of 123 bp. On the other hand, the genotyping of resistance to BGYMV was carried out with the codominant MM SR2 linked to the bgm-1 gene according to Miklas et al. (2005); resistant homozygotes (bgm-1/bgm-1) amplified a 530 bp fragment and susceptible ones a 570 bp fragment. All PCR products were resolved by electrophoresis in 1% agarose and visualized with biotin (GelRed™) under UV light.

For each of the genotypes, the percentage of plants with the MM linked to the resistance genes I, bc-3 and bgm-1 was determined by dividing the number of plants with the MM between the number of plants genotyped and multiplied by one hundred. Plants that did not show symptoms when inoculated with BCMNV NL-3 and had both MM linked to genes I and bc-3 were considered resistant to BCMNV and BCMV; while those that did not show symptoms when inoculated with BGYMV-MX and had MM linked to the bgm-1 gene were considered resistant to BGYMV.

Phenotyping resistance to BCMV, BCMNV and BGYMV

With the exception of the Negro 17-99 genotype, from which no seed germinated, the resistance to BCMV and BCMNV of 12 plants of 50 genotypes was phenotyped, in the remaining 19 genotypes between 8 and 11 plants were evaluated. The 797 plants were inoculated corresponding to 69 genotypes with strain BCMNV-NL-3 and seven days later, 631 plants showed localized necrotic lesions (LNL) in the primary leaves inoculated (Figure 1F), which indicated the presence of gene I and absence of some recessive gene bc (^{Drijfhout, 1978}). Nine days after the re-inoculation some plants developed LNL, and 21 DDI, common mosaic symptoms were observed in 120 plants (Figure 1I and 1J) and no symptoms in 46 plants (Figure 1H). The plants that presented LNL at the first inoculation developed black roots and died (Figure 1G).

The symptoms of LNL usually manifest from 6 to 10 DDI, while those of common mosaic take between 15 and 20 DDI (^{Drijfhout, 1978}). The development of common mosaic symptoms in plants inoculated with BCMNV NL-3 indicated the absence of gene I, while the absence of symptoms suggested the presence of a recessive gene such as bc1 ² , bc2 ² , bc-3 alone or in combination with the gene I (^{Drijfhout, 1978}).

To phenotyping resistance to BGYMV, between two and nine plants were used with the exception of the genotype Negro 17-99, and the LR 'Negro Citlali'/XRAV-187-3-1-8 in which there was no germination. The inoculated plants showed diverse symptoms that included the absence of symptoms (Figure 1C), yellowing of ribs (Figure 1A), yellow gold mosaic (Figure 1B), leaf deformation (Figure 1E), shortening of ribs (Figure 1D) and in cases where the infected plants were allowed to reach stage R8, flower abortion and pod deformation were observed.

Figure 1 Phenotyping resistance to BCMV, BCMNV and BGYMV in the genotypes of the assay. A, B, C, D and E: symptoms induced by the inoculation of BGYMV-MX. F, G, H, I and J: symptoms induced by BCMNV NL-3. A) yellowing of ribs; B) golden mosaic; C) dwarfism in a susceptible plant (left) compared to a resistant plant with the bgm-1 gene (right); D) shortening of internodes and abortion of flowers and pods; E) leaf deformation; F) localized necrotic lesions in the genotypes that only possess the I gene; G) black root; H) resistant plants with the genetic combination I + bc-3; I) mosaic in ribs; and J) common mosaic. 

By PCR, the presence of BGYMV was confirmed in the plants that developed symptoms, while in the asymptomatic plants no amplification was obtained, which confirmed the absence of the virus. In bean agroinoculation works with the BGYMV-MX strain, ^{Garrido-Ramírez et al. (2000)} observed differences according to the origin of the genotypes. It is probable that the variety of symptoms expressed in the inoculated plants was due to the fact that the LRs have DOR-454 and PR0003-124 in their pedigree, so that in addition to the bgm-1 gene, some could also have the QTL of greater resistance (SW12), which would explain the heterogeneity in the development of symptoms.

Genotyping with molecular markers linked to genes I, bc-3 and bc1 ²

With the intention of reducing the cost of detection of the MM linked to the resistance genes to BCMV and BCMNV, a sample composed of all the plants of each genotype was genotyped. The electrophoretic pattern of MM SW13, ENM and SR2 is shown in Figure 2. In all composite samples, MM SW13 linked to gene I was detected. MM SBD5 was detected only in the LR ‛Negro Papaloapan’/SEN-46 -6-5, which suggested that at least one plant in this LR possessed the bc1 ² gene, since 10 of the plants developed black root and two necrosis in the stem; these last plants were genotyped individually, and 45 DDI the only plant that had the combination of genes I + bc1 ² died due to the development of black root. The combination of the I + bc1 ² genes confers resistance to BCMNV NL-3 between 17 and 26 °C, but breaks at 30 °C (^{Drijfhout, 1978}); The development of necrosis in 10 DDI veins has also been observed with strain BCMNV NL-3, although it is not indicated at what temperature (^{Bello et al., 2014}). In this work, the temperature oscillated between 30 and 36 °C in the hottest hours of the day.

Figure 2 Representative figure of genotyping with molecular markers linked to resistance genes to BCMV, BCMNV and BGYMV in recombinant black bean lines. Lanes 1 and 2: negative and positive control, respectively. Lanes 3 to 10: problem samples; A) molecular marker (MM) SW13. A 690 bp fragment indicates the presence of MM linked to gene I; B) MM ENM. The presence of two fragments (418 bp and 123 bp) after digestion with the enzyme Fau I indicates the presence of MM linked to the bc-3 gene and C) molecular marker SR2; a fragment of 530 bp indicates the presence of MM linked to the bgm-1 gene. 

In 55 of the genotypes, codominant MM type CAPS ENM was not detected, which indicated that none of them had the bc-3 gene; only 10 of the LR and elite genotype SCN-2 showed the typical banding pattern of a heterozygote eIF4E ² /eIF4E ³ and four that of a homozygote eIF4E ² / eIF4E ² indicating that some LR are still heterogeneous in relation to the presence of this gene.

The plants of each genotype that did not develop symptoms of common mosaic or black root after the second inoculation were genotyped individually with MMs linked to genes I, bc-3 and bc1 ² . In the LR ‛Negro Papaloapan’/SEN-46-4-9, ‛Negro Papaloapan’/SEN-46-7-6, ‛Negro Papaloapan’/SEN-46-7-7, ‛Negro Citlali’/XRAV-187 -3-14-7, ‛Jamapa Plus’/XRAV-187-3-4-1, ‛Negro Papaloapan’/SEN-46-5-5 and ‛Negro Citlali’/XRAV-187-3-1-9 detected between one and three asymptomatic plants inoculation with BCMNV NL-3 that had MM SW13 bound to gene I, but not MM SBD5 bound to bc1 ² gene or MM ENM bound to bc-3 gene, so it was not it was able to determine if the observed resistance was due to some other resistance gene not yet described, to an error in the process of detecting MM or the inoculation of these plants in particular.

Genotypes where plants were detected only with MM SW13 were ‛Negro Papaloapan’/SEN-46-4-5 (one plant), ‛Negro Papaloapan’/XRAV-187-3-1-8 (one plant), SCN- 2 (one plant) ‛Jamapa Plus’/XRAV-187-3-4-1 (one plant) and T-39 (four plants), so they were expected to develop symptoms of necrosis by inoculating them with strain BCMNV NL-3. However, they only presented mosaics in the trifoil. This situation could be due to the fact that MM SW13 can identify false negatives, as reported by ^{Bello et al. (2014)} who developed a MM of more codominant type that theoretically solves this situation.

A plant from each of the genotypes ‛Negro Papaloapan’/SEN-46-7-6, ‛Negro Papaloapan’/SEN 46-7-13 and ‛Jamapa Plus’/XRAV-187-3-4-4 had only the MM ENM in a homozygous state (eIF4E ² /eIF4E ² ), while 15 genotypes had between one and 11 individuals with the combination of genes I + bc-3 (Table 2), of which six stood out for their greater proportion of plants with this combination genetics: ‛Negro Papaloapan’/SEN-46-7-10 (92%); ‛Jamapa Plus’/XRAV-187-3-4-4 (88%); ‛Negro Papaloapan’/SEN 46-5-5 (82%); ‛Negro Papaloapan’/SEN-46-7-7, ‛Negro Citlali'/XRAV-187-3-1-9, and SCN-2 (75%); the other nine genotypes had between 8% and 67% of plants with both genes (Table 2). The varieties and LR with the highest proportion of plants with the genetic combination I + bc-3 represent sources of resistance to BCMNV and BCMV in the bean stock of black opaque type. It is important to note that in some plants with the combination I + bc-3 weak symptoms of what appear to be mosaics were observed. However, it could be due to some nutritional deficiency.

Although the breakage of resistance of the combination of the I + bc-3 genes by the effect of some necrotic strain such as BCMNV NL-3 has not been reported so far and the studies in which the resistance to “high temperatures” have been evaluated were carried out at 32 °C (^{Pasev et al., 2013}), if there are records that the resistance conferred by gene I to strain BCMV NY-15 in the Black Turtle differential variety of GR 8 (Drifjhout, 1978) is dependent on the dose of the gene (II, Ii or ii) and breaks at temperatures above 34 °C (^{Witmer-Collmer et al., 2000}). ^{Drijfhout, (1978)} also reported the breakdown of the resistance of the varieties Black Tourtle Soup (I), Widusa (I), Jubila (I + bc1), Impr. Tendergr. (I + bc1), Top Crop (I +bc1) and Amanda (I + bc1 ² ) when inoculated with strains NL-3, NL-5, and NL-6 at 30 °C, and some of them also developed necrosis systemic with strains NL-2 and NL-8.

Genotyping with the molecular marker SR2 linked to the bgm-1 gene

In some plants of a group of eleven LR: seven of the cross ‛Negro Papaloapan’/SEN-46 and four of ‛Negro Citlali’/XRAV-187-3, and in the varieties DOR 448, ‛Negro Papaloapan’, ‘Verdín’ and T-39 no symptoms were observed, nor was BGYMV detected by PCR. In these plants a single amplicon of 530 bp was obtained, indicating that they were resistant homozygotes (bgm-1/bgm-1). However, in another group of five LR and the varieties ‛Negro 8025’, ‛Negro Veracruz’, ‛Negro Huasteco-81’ and ‛Negro Comapa’, symptoms of BGYMV infection were observed despite having the MM SR2 bound to the bgm-1 gene.

Although in practical terms the bgm-1 gene is the most important for conferring resistance to BGYMV (^{Beebe et al., 1995}; ^{Singh et al., 2000}), MM SR2 linked to this gene is associated with resistance to general symptoms of the disease as yellowing or chlorosis, but not with symptoms such as dwarfism, which is related to other resistance factors such as the Dwf gene (^{Blair et al., 2007}). Other resistance genes such as Bgp1, linked to the normal development of the sheaths during infection (^{Acevedo-Román et al., 2004}), or the W12 locus of resistance to BGYMV (^{Miklas et al., 1996}, ^2000b) are more effective when they are present with the bgm-1 gene.

The DOR-364 line, with the QTL of resistance to BGYMV, significantly increased the resistance to this virus with the incorporation of the bgm-1 gene recovered from the Mexican accession G2402 (‛Garrapato’). Subsequent lines such as DOR-482 had a disease index of 2, on a scale of 1= immune to 9= totally susceptible, while DOR-364 received a rating of 4 (^{Miles and Pandey, 2004}). Because in the present study only MM SR2 was used, it is likely that the absence of symptoms in the LR and resistant varieties is due to the joint action of the bgm-1 gene with some other genes, such as the QTL greater resistance to BGYMV present in the pedigree of ‛Negro Papaloapan’ and XRAV-187-3.

All LRs had a high proportion of plants with gene I, which confers resistance to the non-inducing strains of BCMV necrosis; six LR derived from the cross ‘Negro Papaloapan’/SEN-46 and two LR from each of the crosses ‘Negro Citlali’/XRAV-187-3 and ‛Jamapa Plus’/XRAV-187-3-4-1 had between 8% and 92% of plants with the genetic combination I + bc-3,, which confers resistance to BCMV and BCMNV; whereas the genes I and bgm-1, which confers resistance to BCMV and BGYMV, were detected in seven LR, three of the cross ‘Negro Papaloapan’/SEN-46 and four of the cross ‛Negro Citlali’/XRAV-187- 3, some of which were homogeneous in the percentage of plants with both genes (Table 2).

Four LR derived from the cross ‘Negro Papaloapan’/SEN-46: ‘Negro Papaloapan’/SEN-46-7-8, ‘Negro Papaloapan’/SEN-46-7-11, ‘Negro Papaloapan’/SEN-46- 7-12 and ‘Negro Papaloapan’/SEN-46-7-13 stood out for the percentage of plants with the I, bc-3 and bgm-1 genes (Table 2) that confers resistance to BCMV, BCMNV and BGYMV.

The heterogeneity of the LR in the proportion of plants with each of the resistance genes highlights the need to homogenize them by using MM to develop opaque black bean varieties with general or specific resistance to three of the viruses that most affect bean production in Latin America. The use of varieties with specific resistance to the predominant viruses in each region will contribute to reduce the losses and spread of these viruses within and between producing regions.